Introduction to farm surveys

Transcription

1 Introduction to farm surveys I l

2

3 m m Hl International Institute for Land Reclamation and Improvement/ILRI P.O. Box 45,6700 AA Wageningen, The Netherlands 1986.

4 The work of Dr. Murphy has been financed by the Upper Volta Office of the United States Agency for International Development under contract PSC International Institute for Land Reclamation and Improvement ILRI, Wageningen, The Netherlands This book or any part thereof must not be reproduced in any form without the written permission of ILRI. Earlier edition by International Institute for Land Reclamation and Improvement (ILRI), Wageningen, The Netherlands. ISBN Printed in The Netherlands.

5 HOW TO USE THIS BOOK This book is intended for enumerators with a level of education of some years of secondary school. The book can be used in three different ways: A. In independent study by the enumerators themselves; B. In a training course that prepares the enumerators before they start their survey work; C. In on-the-job training sessions for enumerators whilst they are engaged in surveys. A Independent study The text of the book will be understandable to an enumerator with some experience in survey work. For independent study, the student is advised to read the chapters in their numerical order because they start with simple matters and move on to more complex ones as the book progresses. At the end of each chapter is a set of exercises or questions which allow the student to check whether he has understood the contents of the chapter. He can find the answers to the calculations at the back of the book. The student will find it helpful to discuss the subjects he has studied with his colleagues, his superiors, and with extension agents. V

6 B Training course before a survey In a training course, the enumerators in a project are brought together to be trained before they commence their work. During the course, the objectives of the survey and the work the enumerators will be required to do should be explained to them. This can best be done on the basis of an 'Enumerator's Manual' --a document specially written for each survey program to be undertaken. This manual should describe, step-by-step, the work that will be done (how to fill in the survey forms, how to solve problems that may arise during the surveys, and so on). The teacher can base his course on that manual and che present book. Training before the surveys begin C On-the'-job training sessions At regular intervals, say once a month, the enumerators in a project (or in one region of a large project) can be brought together for a one-day training session. They should be requested to read one chapter of this v1

7 book (or part of one chapter) at home before the training session, and to try to do the exercises and answer the questions listed at the end of the chapter. During the training session, the teacher should check whether the enumerators have understood the pages studied, should explain any point that was not clear to them, and should discuss the solutions to the exercises. He should explain how the knowledge they have acquired from the chapter can help them in understanding their work and how it fits in with their work program. Some chapters are more difficult or longer than others. If necessary, they can be split up and treated in more than one session. Others can be combined. The order of the chapters can be changed to fit the needs of the project, and especially to coincide with the crop calendar. For example: Chapter 10 explains how to measure fields. If the enumerators will be required to do this work shortly after training begins, it may be best to study Chapter 10 before the other chapters. v11

8 CONTENTS A B C HOW TO USE THIS BOOK Independent study Training course before a survey On-the-job training sessions V V vi vi 1 1. I 1.2 INTRODUCTION Purpose of this book Background of the book Chapter 1 : Questions ORGANIZING FARM SURVEYS The need for information The role of a monitoring and evaluation unit The role of the enumerator Chapter 2 : Questions INTRODUCING THE SURVEY Village meeting Visiting the farmers in the sample Building up a good relationship with the farmers Chapter 3 : Questions DATA-GATHERING METHODS Observations and interviews Single and repeated interviews and observations Vlll

9 Open-ended and multiple-choice questions The wording of questions Chapter 4 : Questions OBTAINING THE DATA The survey forms An 'Enumerator's Manual' A regular schedule of interviews Recording understandable data Visiting the fields Surveys of long duration Chapter 5 : Questions SELECTING A SAMPLE Population and sample A representative sample Random sample Stratified random sample Size of the sample Generalizations from sample data Chapter 6 : Questions SUMMARIZING THE DATA General Tabulation Averages Dispersion Chapter 7 : Questions FREQUENCY DISTRIBUTIONS Frequencies Relative frequencies Graphic presentation of frequency tables Chapter 8 : Questions RELATIVE CHANGES General x

10 Constant increase Ratios Graphic presentation of changes over time Chapter 9 : Questions IO IO MEASURING FIELDS Geome try Measuring distances Measuring angles Measuring a field Chapter IO : Questions ESTIMATING CROP PRODUCTION The use of scales Direct estimates of crop production Indirect estimates of crop production Chapter 11 : Questions SOURCES OF ERROR IN DATA Sample errors Errors due to instruments Human errors Rounding off Chapter 12 : Questions O NATURAL CONSTRAINTS ON FARMING C 1 ima te Characteristics of the soil Soil erosion Chapter 13 : Questions O FARMING PRACTICES Land preparation Sowing Crop maintenance Application of fertilizer Crop protection 1 o9 1 o I I6 X

11 Cropping pattern Crop rotation Chapter 14 : Questions 1 I7 1 I AGRICULTURAL ECONOMICS Introduction Some terms used in agricultural economics Costs and investments Costs in cash and in kind Direct and indirect costs Credit costs Cost of investments Labour unit Chapter 15 : Questions NET PROFIT FROM CROP CULTIVATION Gross margin of a crop Gross margin per hectare Gross margin per man-day Gross margin per unit of cost Gross production value Net profit from crop cultivation Chapter 16 : Questions OTHER HOUSEHOLD INCOME Net profit from livestock Net profit from trade and other activities Chapter 17 : Questions CONSTRAINTS ON CHANGES IN FARMING PRACTICES Physical environment Technical feasibility Economic return Social acceptability Conclusions Chapter 18 : Questions x1

12 ANSWERS TO THE QUESTIONS Chapter 6 Chapter 7 Chapter 8 Chapter 9 Chapter 10 Chapter 11 Chapter 12 Chapter 15 Chapter 16 Chapter i x11

13 INTRODUCTION 1. 1 Purpose of this book We have written this book for enumerators who will be trained to work in farm surveys. Their work will be to collect information from farmers about.how they farm and about the results they achieve. The book presents basic background information on farm surveys, mathematics, farm management, and farming practices - matters that an enumerator should understand if he is to do his job properly. The purpose of this book is to show enumerators why they are asked to gather information, how it will be used, and how the quality of their work affects the results of the farm surveys. Enumerators who understand the reasons for the surveys are more likely to be accurate and thorough in their job, and will keep on working steadily and carefully over a long period of time. They will also be better able to explain their work to the farmers and encourage the farmers to cooperate. 1.2 Background of the book This book describes the work of enumerators among small farmers in semiarid West Africa. This is an area with one rainy season lasting from four to eight months. The total rainfall is between 400 nun and 1,200 m, so rain-fed farming (farming depending only on rainfall) is possible. Irrigation (bringing water to the crops) cannot be practised because

14 water is not available in large quantities. Rainfall in Africa Most farms in semi-arid West Africa are family farms, with the farmer and the members of his household working together to produce food crops. The household consumes most of the production itself. The small part remaining is sold to pay taxes and cover other cash expenditures. This type of farming is called subsistence farming. Most of the land cultivated is used to grow food for the household's own consumption; only small areas are used to grow crops which can be sold. n L

15 If a crop is cultivated to be sold, it is called a cash crop. Examples are cotton and peanuts. Because there are great differences in the amount and timing of the rains from year to year, the quantity of crops harvested varies greatly. Sometimes in bad years, some households do not produce enough food to feed themselves. Even in good years, the surplus is likely to be small. Traditionally, subsistence farmers plough their land with hand tools, sow local varieties of seeds, and apply little or no fertilizer to their crops. They sometimes use animal traction (a donkey or an ox). In this kind of traditional farming, farmers grow the same type of crop on the same field for many years. When the land is exhausted and the harvest becomes very small, the field is left fallow. This means that it is abandoned to the natural vegetation. After some years, it can be cultivated again. This system of farming works well as long as populations remain small and land remains plentiful. In many parts of semi-arid West Africa, however, populations nowadays are large and unoccupied land is scarce. 3

16 Chapter 1 : Questions (All pertaining to the area you are working in): I. In what period of the year do the rains fall? 11. About how much rain falls: In a normal year? In a wet year? In a dry year? How much rain fell last year? 111. In a normal year, do the farmers produce enough food for their households? What about the country as a whole; can it produce enough food for its people? IV. Can a farmer acquire more land if he needs it? V. How many years in a row do the farmers cultivate the same field before putting it in fallow (if a fallow period exists)? 4

17 2 ORGANIZING FARM SURVEYS 2.1 The need for information It is vital that farmers increase their food production. Populations are growing and everywhere there is a demand for more food. If farmers can grow more food than they need for their households, they can sell the surplus. Large surpluses can even be sold to other countries. A country can increase its food production in two ways: it can farm more land and it can get bigger harvests from the land already being farmed. Throughout the world, agronomists are developing new, higher-yielding varieties of seed, and everywhere farmers are being encouraged to use new farming techniques that give better results than their traditional farming practices do. The new farming techniques include the protection of plants from diseases and pests, the conservation of the soil, the use of animal traction, and the use of manure and chemical fertilizers. These techniques are usually developed and tested at agricultural research stations, where 'farming' conditions are very good. At a research station, all the 'farming' operations are technically correct and they can all be done at the proper time. Money and labour are in relatively plentiful supply at a research station. It is therefore difficult to be sure whether the techniques developed there will be as good when they are used by a real farmer. 5

18 In real life, farmers cultivate a fairly large areafwith only a small number of workers (the members of their households). They have many social obligations to fulfil and need to go to the market or to visit relatives once in a while; they cannot spend all their time farming. Sometimes, they cannot sow all their crops at the proper time; other times they will weed some crops poorly or too late; often, they will not have much money to spend on fertilizer, so they use only small quantities, if any at all. Farmer applying fertilizer Even if a farmer is using new farming techniques, all these elements will influence the results he really achieves. In some cases, if he does not fulfil a set of minimum requirements - like applying a minimum amount of fertilizer or doing a minimum amount of weeding - the new farming.techniques may cost him more money than they are worth, because they do not pay off in bigger harvests. Whether farmers are following traditional techniques or applying new ones, the agronomists working at research stations need to know the results really being achieved by the farmers. If they have this in- 6

19 formation, they can modify and adapt the new farming techniques and make them more suitable to the farmers' real conditions. The people who organize programs of agricultural extension also need to know what the farmers are achieving. If they know this, they can see whether their programs are really helping the farmers; if not, they can decide what changes they should make in their future programs. Theoretically, if an agronomist or the head of an extension service wants to know about the results being achieved by farmers, he could go to their villages and talk to them himself. But this would take up a great deal of his time and would probably conflict with the rest of his work. Besides, he would not be able to talk to enough farmers or to visit enough,villages. So, instead, a farm survey is conducted. In a farm survey, information is obtained from a great many farmers. The persons who obtain this information are called enumerators. The information they collect is then passed on to agronomists or to the heads of extension services. Farm surveys focus not only on agriculture; they also cover aspects of the social life in the villages, because these greatly influence the changes that the farmers are willing to make in their farming practices. For example: If it is known that the farmers spend only 60 per cent of their days on farming during the rainy season, one cannot automatically conclude that more labour-intensive farming practices could be introduced. It is quite possible that the farmers spend the remaining 40 per cent of their days on social activities that they consider more important than a bigger harvest! I farmers. 7

20 For example: If the agricultural extension office wants to introduce a cash crop in a region, it first needs to know whether the farmers are already producing enough food for themselves, and are therefore able to reserve part of their land to grow cash crops on. If not, the extension program can far better start helping the farmers to grow more food before considering the introduction of cash crops. Farm surveys are also very useful during the course of extension programs; they can show whether the farmers are applying the recommended techniques and, if not, why not. The surveys can also reveal whether the farmers who are applying the new techniques are obtaining the results that were expected. 2.2 The role of a monitoring and evaluation unit The managers of large development projects often decide to employ enumerators on a permanent basis. The enumerators then work in the project in a special unit which organizes and coordinates the farm surveys. This unit may be called a 'Statistical Unit' or a 'Monitoring and Evaluation Unit'. Monitoring means that the staff of the unit continuously observe the activities of the farmers; evaluation means that these observations are analyzed and studied. A monitoring and evaluation unit can be attached directly to the general management of a project, or it can be part of an extension service or an economic planning service. The unit will receive requests for information from various specialists who need this information for their work. The director of the unit then combines these requests and designs a program of farm surveys to collect the information that has been reque s t ed. The director of a monitoring and evaluation unit is usually an agricultural economist. He organizes the surveys, supervises the work of his staff, and pres.ents the results in reports. The staff includes enumerators (who collect the information) and data processors (who group and tabulate the information collected by the enumerators). The unit may 8

21 also have its own secretary, although in most cases it may have to share a secretary with others in the project. office Goorker Staff of a monitoring and evaluation unit If a unit has many enumerators, it will also have a number of supervisors who assist the director in the field. Each supervisor is responsible for a group of enumerators. He guides them in their work and sees to it that they receive the necessary equipment to do their job. The supervisors report regularly to the director on how the work is going in the field. When a survey is to be conducted, the director of the unit consults with 9

22 the people who are requesting information. Together, they define the precise objectives of the survey; this means that they define precisely what information is to be collected. The director then decides how it will be collected and prepares the survey forms which the enumerators will use to write down the information they collect. The director will also consult with the enumerators while he is pre- paring the surveys. This will be discussed in Chapter The role of the enumerator Each enumerator should understand that the quality of his work greatly affects the results that his unit is capable of producing. The results of a survey will be useful only if the information that is collected is valid. This means that it must be accurate and complete, and that it must be collected at the proper time. The work of an enumerator consists of three elements: - He has to conduct interviews and make observations as instructed, and has to note the farmers' answers and his own observations accurately; - He has to behave in such a way that the farmers feel at ease with him and are willing to tell him the truth; - He is constantly observing and comparing his observations, so that he notices if a farmer is forgetful or not entirely truthful, and so that he can report to the unit's office any unusual situation which helps explain the information he has collected. All the efforts of enumerators are wasted if the farmers are not willing to cooperate. The fact that the farmers readily answer questions does not mean that they are telling the truth; they may just be trying to get rid of the enumerator. It is therefore important that the farmers understand from the start what the purpose of the survey is, what they will be asked to do, and what the results will be used for. 10

23 Chapter 2 : Questions (All pertaining to the area you are working in): I. Describe the farming practices of the farmers in your area. 11. What new farming techniques are being introduced in your area? 111. Can you think of any aspects of traditional life that might make the farmers reluctant to adopt (new) techniques such as: - Animal traction? - Improved seed varieties? - Cash crops? - The use of fertilizer and insecticides? - More labour-intensive farming practices? 11

24 ~ 3 INTRODUCING THE SURVEY Before a survey can begin - while it is in its preparatory phase - the director of the unit will contact the proper authorities within the administration and within the traditional hierarchy. He will then explain to them the reasons for the survey. The attitude of the authorities towards the survey will influence the attitude of the farmers, so care must be taken not to antagonize the authorities when introducing the survey. If they understand the reasons for the survey and are in favour of it, they can provide valuable information about the history and the infrastructure of the area. They can also supply useful information about the villages that might be selected for the survey. The procedure for introducing the survey to farmers in the villages depends partly upon local customs, and especially upon the influence of the traditional chief. He is an important man and his behaviour towards the director and other members of the unit's staff will greatly influence the behaviour of the farmers, so it is vital to gain his cooperation. Whatever procedure is followed, it is important to keep everything out in the open. Acting in this way will help to prevent the spread of rumours (that the information to be collected will be used for tax purposes or that the farmers who will take part in the survey have been selected for special treatment). The survey should be introduced to the farmers in two stages: first, to

25 ~ all the farmers in a village, and second, individually to each of the farmers who will take part in the survey. 3.1 Village meeting After explaining the survey to the traditional chief, the director of the unit will ask him to call a meeting of the villagers. During this meeting, the chief will announce the survey and will introduce the 'I director, the enumerator who will work in the village, and the enu- merator's supervisor, if any. This will show that the chief is informed about the survey and agrees with the need for it. A The traditional chief welcomes the director and the enumerator to the village meeting The director should then explain in simple terms which agency they are working for and what the objectives of the survey are. These objectives should be presented in general terms. For example: We want to find out if the project is really achieving the results- that were expected.

26 Or: We want to find out the best way to grow peanuts in this area. Immediately after that explanation, the director should describe the steps involved in the surveys in very concrete terms. For example: We will measure all the crops at harvest time right in each farmer's compound and we will not take anything away. Or: We will note, step by step, how the farmers here grow their peanuts and then measure the harvest so that we can all understand which is the best way to grow peanuts. This explanation should be quite detailed, emphasizing especially the physical handling of the crops when harvest estimates are being made. During the meeting, the farmers should be informed about the work of the enumerator so that they will know what to expect. For example: The farmers should be told where the enumerator will live, how often he will visit the farmers' compounds, that he will observe their work in the fields, that he will write down his observations and the information given to him by the farmers on forms which he will later send to the unit's office. It should be emphasized at the meeting that those forms will be sent to the office so that the results can be calculated and compared with those of other farmers, but that the forms will not be given to the government for tax purposes or for any other reason. At that stage, it is easy to explain that it would be impossible for the enumerator to do this work with all the farmers, and that therefore a group of farmers will be chosen to represent the village. It is important to insist that the farmers who will be selected are neither better nor worse than the others, and that they will be neither punished nor particularly rewarded for taking part in the survey. Whenever possible, the farmers who will take part should be selected at the meeting. One way to do this is to write the names of all the farmers 14

27 on small pieces of paper, put these papers into a hat, and ask the traditional chief to take out the required number of papers. Selecting the farmers in this way - out in the open in front of everyone - will make it clear that the selection was not arranged beforehand. The group of farmers so chosen is known as the 'sample'. 3.2 Visiting the farmers in the sample The next stage in introducing the survey is that the enumerator and the director (or the enumerator's supervisor, as the case may be) locate the homesteads of the farmers in the sample and visit each of them individually. The main reasons for this visit are to introduce the enumerator personally to the farmer and the members of his household, to explain to them the work involved in the survey, and to answer any questions the farmer might have been hesitant to ask in front of his neighbours at the village meeting. During the visit, the farmer will once again need to be reassured that his participation in the survey will not create difficulties for him with taxes or with agricultural credit, and that the local authorities know about the survey and are favourable towards it. The enumerator should be aware'that farmers are often genuinely interested in finding out how much food they produce and what their incomes really are (measured in more official units of measurement than the local ones). He should also realize that the farmers are more likely to be truthful in their answers to his questions if they know they will be kept informed of the findings of the survey while it is in progress. The farmers, on their part, should be made aware that if they give inaccurate or incomplete information, they are putting the entire work wf the unit in jeopardy. (In areas with food-aid programs, farmers may systematically under-report their crop yields, hoping for more food aid.) The first visit to a farmer is very important because it lays the basis 15

28 for cooperation between the enumerator and the farmer - cooperation which is essential to the success of the survey. n a Visiting a farmer at his compound At the beginning of a survey, it may happen that a farmer selected in the sample refuses to take part in the survey and cannot be persuaded to do so. It may also happen that one of the farmers selected plans to migrate shortly afterwards. In such cases, other farmers can be chosen to take their places. This will not hurt the survey in any way. It is far more detrimental, however, if a farmer has to be replaced half-way through the crop season, because the 'new' farmer will find it difficult to remember details of what he did early in the season. The enumerator should therefore do everything he can to avoid 'losing' a farmer during the crop season. After the village meeting has been held and the individual visits have been made, it is polite for the enumerator to visit the traditional chief again to thank him and to inform him of any farmers who have been replaced in the. sample. 16

29 3.3 Building up a good relationship with the farmers The behaviour of the enumerator towards the farmers is crucial to the success of the survey, whether he lives among the farmers during a longterm survey or whether he only comes to the village for short visits. The farmers do not need the enumerator; it is the other way around! An enumerator who comes to a village and starts bossing people around to come for interviews is not going to obtain data that are worth processing. An enumerator who is polite but seems in a hurry and just writes down what the farmer says without thinking about it is not going to get good data either. The first task of the enumerator is to make himself accepted by the people in the village. This means behaving politely. The enumerator is a stranger and should behave accordingly, going to greet the traditional chief and the authorities according to custom, and always being polite to everyone. He should also take the time to explain to the villagers who he is and what work he is there to do. His own director (or supervisor) has already introduced him to the authorities and explained the objectives of the survey, but even so, it is likely that further explanations will be needed. The enumerator should always be willing to answer questions about the survey from the authorities and the villagers. Sometimes in the villages, there may be conflicts between families or between neighbourhoods or groups. The enumerator must be very careful not to let himself get involved in these conflicts. When he first arrives in a village, it is easy for him to become involved without even knowing it, simply by accepting an invitation to rent a room in a homestead or a case in a compound, or even by talking often with someone who is very friendly, not knowing that this person may be disliked in the village. Until the enumerator has had time to find out what the situation in the village is, it is best for him not to become too clearly associated with anyone. If the enumerator lives in the village, he should be aware that his 17

30 behaviour and way of life are going to be observed by the villagers, so he must obey the usual rules for strangers to the village and take care that his visits to the various homesteads can never be misinterpreted. (This is especially important if he is not married.) He should dress in a manner appropriate to going into the field; formal city dress is not practical for field work. Even after the enumerator is well accepted in the village, his behaviour during each interview remains important. If a farmer is upset because the enumerator interrupts the farmer's work or rushes through the salutations, the farmer may give only minimal answers to his questions. The enumerator should be particularly careful not to gossip about the families he surveys, and never to repeat any private and personal information which the farmers entrusted him with in the course of his work. That could cause him to lose the trust of the farmers and may even force him to leave the village. Any questionable attitude towards the women of the village could have the same result. Enumerators should realize that a vital part of their work is gaining and keeping - the trust of the farmers. - 18

31 Chapter 3 : Questions I. If you had to introduce a survey in your village, which persons would you approach before starting the survey? 11. a) In your village, have you noticed any conflicts between groups or between families? If so, what are the underlying causes of those conflicts? b) Suppose an enumerator goes to a village to do a survey. What tips could you give him so that he has a better chance of keeping out of conflicts? 111. Suppose a farm survey is to be done in your village. The survey will investigate what results the farmers are achieving in their fields, how much it costs them to cultivate their crops, and what income they earn from activities other than farming. a) Do you think there might be some matters on which the farmers may not want to cooperate? If so, what are these matters? b) How do you think this cooperation could be obtained? c) What would you do to keep the farmers' cooperation in a survey that lasts several years? 19

32 4. DATA-GATHERING METHODS 4.1 Observations and interviews The data that an enumerator will be required to gather during a survey can be obtained in two different ways: - By observing what is going on in the fields; - By interviewing the farmers. For example: After animal traction has been introduced in an area, the extension service will want to know whether the farmers are indeed using animal traction to plough their fields. An enumerator can be sent to observe the farmers during ploughing to see whether they plough by hand or with animal traction. This is data-gathering by observation. The enumerator can also be requested to go to the farmers and ask them whether they are using animal traction or not. This is datagathering by interviews. If the farmers tell the truth, it is much easier and takes less time to ask them - even months after the event - than to follow them about to see what they do. Besides, many events are too complex to be studied by observation, or the survey would become too expensive because it would require too many enumerators. An enumerator can handle only a limited number of events. For example: Labour-time studies by observation could only be done if 20

33 each enumerator followed one or two families at the most, or observed only a few fields located near one another. Enumerator observing the development of a crop Information obtained by interviewing is likely to be less accurate than that obtained by observation because the farmers may not remember or do not wish to tell. On the other hand, the quantity of information that an enumerator can gather by interviews is much greater; so interviewing is much cheaper per item of information obtained. In practice, it is common to combine interviews and observations. In this way, accurate data are obtained (by observations) and at low cost (by interviews). For example: If one wants to know how much of their crop the farmers in a village sell at the local market or elsewhere, one could observe them to see what they do. That would be very expensive because,many enumerators would be needed. It is much easier to visit each farmer regularly and ask him whether he has sols part of his crop, even though he might not always tell the truth. A combination of regular interviews and regular visits to the local market makes it possible to check part of the information obtained from the farmers. A farmer who says he did not sell anything although he was seen at the market 21

34 selling a crop might be telling the truth; perhaps he was selling it for a friend. The enumerator should inquire carefully whether it was the farmer's own crop that he was selling or someone else's. This survey thus combines interviewing the farmer with direct observations at the market. 4.2 Single and repeated interviews and observations If a survey covers an event that has passed and that happened only once, it is enough to interview each farmer once only (e.g. asking him about his harvest of the past year). One interview is also enough if the purpose of the survey is to find out about a situation at a certain point in time. For example: Suppose data are required about the number of people in the household, the type of crop being cultivated, and the tools that are being used to cultivate the crop. A farmer can provide all that information for one crop season in one interview. If the purpose of the survey is to study activities that are taking place over time, the interviews have to be repeated at regular inter- vals. 1 For example: If the trading activities of a member of the household are to be studied throughout the year, the trader will be asked questions every one or two weeks, depending on how much trading he or she does. With some activities that take place over time, the survey may only require information about the end results of that activity. If so, one interview just after the activity has been completed will suffice. For example: The costs and benefits of cotton growing are to be studied in a region. It is known that in that region cotton is grown by household labour only'and that the entire cotton crop is sold to 22

35 the government marketing board. Here, it is appropriate to interview the farmer just after the sale of his cotton. He will easily remember how much it cost him to grow the cotton crop because the few inputs he had to buy are relatively expensive. And he will have no trouble remembering how much money he received from the marketing board. If the survey requires detailed information which the farmer cannot possibly remember for a long time, he will have to be interviewed often. For example: If information is required about the work done by each member of the household or about the quantities harvested day by day, the farmer will have to be interviewed often because he cannot be expected to remember those things for more than a few days. This is in contrast to many other activities such as selling a sheep or applying fertilizer, which will be remembered clearly for several weeks. In a survey with frequent repeated interviews, it is usually arranged that the enumerator lives in the village of the farmers he will visit. This enables him to obtain more detailed and more precise information than if he lived somewhere else and only visited the farmers now and again. Living in the village also makes it easy for the enumerator to visit the fields to make observations. These may be either single observations (e.g. to see what type of crop is being grown on a certain field) or repeated observations (e.g. to see how a crop is developing). 4.3 Open-ended and multiple-choice questions An enumerator can obtain information from a farmer by conversing with him, just as two neighbours will do. In this way, however, he might not obtain all the information he wants because, in a conversation, he is likely to forget some of the topics that he should talk about. Besides, later at the office, the data processors have to tabulate together the information obtained by all the enumerators. This would be impossible 23

36 for them to do if each enumerator has a different conversation with each farmer. For these reasons, the questions that the enumerators will ask should be written down in advance. These questions can be presented to the farmer in two different ways: - As open-ended questions; - As multiple-choice questions. Open-ended questions are printed on the survey form and are read to the farmer exactly as they are written and his answer is copied on the form exactly as he says it. Open-ended questions make a survey rather timeconsuming, so they are only practical for a single-visit survey or a reconnaissance survey. (A reconnaissance survey obtains basic information on which a more in-depth, long-term survey will be designed.) Interviews are held wherever convenient for the farmer With multiple-choice questions, the questions are printed on the survey form, along with a list of their possible answers. The enumerator reads 24

37 ~ a question to the farmer, who answers in his own words, and the enumerator marks on the form the answer that best covers what the farmer has said. The enumerator can also read the question and its list of possible answers and have the farmer choose the right answer, which the enumerator then encircles or marks with a cross. Example of a multiple-choice question Which of the following cereals did you grow this year? (Mark the 1 correct answer with a cross.) I I I 1. White sorghum 2. Red sorghum 3. Millet 4. Maize 5. Rice 6. Others... Note that multiple-choice questions should always have an answer 'Others' so that any unusual answer can be written down. Multiple-choice questions have several advantages: they help the farmer remember things he might not think of mentioning himself, and they are easy for the data processors to read, check, and tabulate. In a long-term survey, open-ended questions could be used in the first period. Then, afterwards, the information they provide could be used to compile multiple-choice questions and their lists of possible answers. 4.4 The wording of questions The farmer should only be asked questions that he is able and willing to answer. The questions should be worded in such a way that they are easy for him to understand and simple for him to answer; the questions must be clear. Also, the questions should not upset him or influence his 25

38 answer; the questions must be neutral. A question should not mention units that a farmer does not understand. He will then not be able to answer even if he wants to. For example: If a farmer is asked: 'How many days does it take you to plough a hectare?', he cannot answer if he does not know how much land a hectare represents. It is also possible that a farmer does not want to answer a certain quest ion. For example: If he is asked: 'How much money did you spend to buy dresses for your first wife?', he may know the answer but does not want to tell it because he considers the matter too private. These example are very obvious, of course, but all questions should be looked at to see whether the farmers can answer them truthfully and are willing to do so. Each question should be worded in such a way that it is clearly understood by both the enumerator and the farmer, with no possible misunderstanding about its meaning. This is quite difficult to do because what seems clear to one person is not always clear to another. Sometimes a question is too broad or too complex; it is, in fact, several questions in one, and the farmer will not know how to answer. Before the questions are finally printed on the survey forms, therefore, it is a good idea to ask several people, including some farmers, what they think each question means. Some very common words, such as 'family' or 'field', have more than one meaning, so it should be explained clearly which meaning is used in the quest ion. For example: A question about a farmer's 'family' could mean all those people with whom he has blood or marriage ties, or it could mean his wives and children only. 26

39 i The way in which a question is worded should not influence the farmer's answer. He should never feel that one answer is better than another, or that his answer could get him into trouble or could displease the enumerator. The enumerator should never show that he thinks the farmer has done something wrong. For example: If the farmer is asked: 'Did you use fertilizer as the extension agent told you to do?', he will feel that it is safer for him to answer 'Yes', even if it is not true. 27

40 Chapter 4 : Questions I. A monitoring and evaluation unit has been created in a project to gather and analyze data about the following topics: Rainf al 1 ; Rotation of crops; Cost of crop cultivation; Labour input; Crop performance; Use of animal traction; Market prices of crops; Income from off-farm activities. a) On what topics would you recommend that data be gathered by observations? b) On what topics would you recommend that data be gathered by interviews? c) On which topics would you recommend repeated interviews or observations? Explain your answers. 11. What do you think is wrong with the following questions? a) Did you protect your seeds against insects correctly, using the prescribed chemical? b) What is the date of birth of each member of your family? c) Do you have draught animals and equipment and do you use them to farm your land? 28

41 5 OBTAINING THE DATA 5.1 The survey forms Before the start of the survey, the director will make a list of the data that are to be gathered by the enumerators. He then decides which data will be obtained through interviews and which by observations. Next he prepares the survey forms that will be used by the enumerators to write down the data they collect. These forms are also called questionnaires. They should be prepared very carefully because they have a great influence on the accuracy of the data gathered and also on the accuracy of the tabulations made at the office afterwards. To prepare the forms, the director first makes a list of all the questions that will be asked and all the observations that will be made. He decides how often the farmers will be interviewed and how often the observations will be made. He then groups the questions into categories, decides which categories can be presented on the same survey form, and prepares a first draft of each form. Excamples of survey forms There may be a form for data on the household (age and sex of the people living in it and the type of work they do). This form can be filled in during the first visit to the farmer. 29

42 There may be another form for the agronomic data on each crop (sowing time, fertilizer used, quantity harvested, etc.). Data will be entered on this form throughout the crop season. Yet another form may be for data on livestock and the income obtained from it. Data will be entered on this form once a week throughout the year. In a long-term survey, it can be useful to group the questions on one topic into a table, which the enumerator simply fills in during each interview. For example: Table 5.1 is a table grouping questions about livestock. Using this table, the enumerator visits a farmer once every week to inquire about the farmer's livestock activities. After the usual greetings and informal talk, the enumerator sits down and first writes the date of that day in the first column on the left. He then asks the farmer if he has bought any livestock since the last interview. Suppose the farmer says: 'No, I have not'. The enumerator writes on the line next to the date: 'None' or some other word to indicate that no livestock has been bought. The next week he comes back to the same farmer and asks the same question. Now the farmer says he bought a goat for 2,000 CFA from his neighbour who needed cash at that time so he got the goat quite cheaply. Table 5.1 shows how this information is recorded. At the end of the month the totals are calculated and discussed with the farmer. The form is then sent back to the office and the enumerator starts a new form, or he continues on the same form for the next month. Of course, this is only part of a livestock survey; other questions will be asked at the same time about sales of livestock and cost of upkeep of the herd. The first draft of the set of survey forms is made by the unit director at the office. He then discusses the forms with the enumerators. This 30

43 Table 5.1. Example form for livestock purchases

44 discussion will be most useful if the enumerators have been living in the villages for a while and so know the farmers and their habits. The matters to be discussed are whether the questions are clear and neutral and whether they are properly worded, how they are presented on the forms, how often the forms will be filled in, and which forms can be filled in during one and the same visit. For example: Sometimes the enumerator can fill in several short forms at one time, asking the farmer about work done in the field and about livestock activities. Director and enumerators discussing the survey forms After the discussion, the director will compile the second draft of the forms and will decide what program the enumerators should follow so that the interviews and observations are done regularly and in the same way by everyone. The programs and the forms should then be tested in the field. The 32

45 enumerators follow the prescribed program and use the forms to interview a few farmers. After a couple of weeks, the director calls another meeting to discuss with the enumerators any problems encountered with the program or the forms. The director will then draw up the final program and the final set of forms. Both program and forms should then remain unchanged for the entire crop season. 5.2 An 'Enumerator's Manual' To do a good job, an enumerator needs to understand the objectives of the surveys and why the results are needed; he can observe well only if he knows what to look for. At the beginning of a survey, the directcr will explain these objectives to the enumerators, and will inform them on such matters as what their work schedules will be, how they should fill in the survey forms, and how they should handle any unusual situations. As an aid to memory, he may provide them with an 'Enumerator's Manual', a document that contains all this information. Such manuals, intended to guide the enumerators in their work, are especially written for each survey program. Enumerators should study their manuals carefully and use them as a reference whenever they are not sure how to record an answer A regular schedule of interviews It is important to set up a regular schedule of interviews with the farmers so that they will know when to expect a visit from the enumerator. If there is a delay, a farmer is more likely to forget some information or to make a mistake. Besides, it is always a problem for an enumerator to catch up with one farmer without delaying his other interviews too. Delays will be kept to a minimum if the enumerator has a good relationship with the farmer, who will then keep the enumerator informed of his travel plans and will make an effort to be at home at the time agreed on for the interview. 33

46 Sometimes, delays in interviews cannot be avoided. Perhaps the date set for the interview coincides with a religious holiday or the farmer is ill. If the enumerator has established a good relationship with the farmers, however, he will be kept informed of events (trips, ceremonies, illnesses) and can rearrange his schedule accordingly. 5.4 Recording understandable data When filling in the survey forms, the enumerator should constantly ask himself: 'Will the people at the office be able to understand what I write down?' This means they must be able to read it (so his writing must be clear and legible) and they must be able to understand it (they must be able to make sense of it). Many times, simply noting the farmers' answers is not enough. The answers are not fully understandable because the situation which caused them is not explained. For this reason, most forms provide space for comments, in which any unusual answers can be explained. Even the most carefully designed survey forms cannot foresee all the circumstances that might affect the data. If the enumerator obtains an important item of information for which there is no space on the form, he should make a note of it in his observation notebook. During the next visit of his director or supervisor, he should bring up the matter and will be advised how to handle it. 5.5 Visiting the fields The enumerator will usually be asked to go regularly to the fields of the farmers he is surveying and observe certain things. To do this work correctly, he must visit all those fields, not just those close to the road or otherwise easy to get to. In the fields, the kinds of things he will be asked to observe are: - Can the crop be expected to yield much or not? - Is the crop damaged by diseases, insects, or animals? 34